water

Article
Study of Sludge Particles Formed during Coagulation
of Synthetic and Municipal Wastewater for Increasing
the Sludge Dewatering Efficiency
Lech Smoczynski 1 , Slawomir Kalinowski 1 , Igor Cretescu 2, * , Michal Smoczynski 3 ,
Harsha Ratnaweera 4 , Mihaela Trifescu 1 and Marta Kosobucka 1
1 Faculty of Environmental Management and Agriculture, University of Warmia and Mazury in Olsztyn,
10719 Olsztyn, Poland; lechs@uwm.edu.pl (L.S.); kalinow@uwm.edu.pl (S.K.);
trifescumihaela@gmail.com (M.T.); marta.kosobucka@uwm.edu.pl (M.K.)
2 Faculty of Chemical Engineering and Environmental Protection, “Gheorghe Asachi” Technical University of
Iasi, 700050 Iasi, Romania
3 Faculty of Food Sciences, University of Warmia and Mazury in Olsztyn, 10719 Olsztyn, Poland;
michal.smoczynski@uwm.edu.pl
4 Faculty of Sciences and Technology, Norwegian University of Life Sciences, 0454 Ås, Norway;
harsha.ratnaweera@nmbu.no
* Correspondence: icre1@yahoo.co.uk or icre@ch.tuiasi.ro; Tel.: +40-741-914-342




Received: 30 August 2018; Accepted: 15 October 2018; Published: 9 January 2019


Abstract: Municipal wastewater sludge was produced by chemical coagulation of synthetic

wastewater (sww) based on Synthene Scarlet P3GL disperse dye and real municipal wastewater (nww),
coagulated by commercial coagulants PAX (prepolymerised aluminum coagulant) and PIX (a ferric
coagulant based on Fe2 (SO4 )3 ). An attempt was made to correlate the sludge’s dewatering capacity
(in terms of capillary suction time—CST) with operation parameters for wastewater treatment, size
distribution and specific surface area of the sludge particles. It was found that the presence of
phosphate ions in the system facilitates the removal efficiency of the above-mentioned dye (L) due
to the interaction between the dye molecules and H2 PO4 − ions. Unlike sww, negatively charged
organic substances (sorg) in nww are directly adsorbed on the surface of colloidal particles {Fe(OH)3 }
and {Al(OH)3 } (prtc). It was also discovered that an increase in the dose of a coagulant led to an
increase of CST for sww sludge and to a decrease of CST for nww sludge. It has been suggested that
flocs composed of spherical {Al(OH)3 } units possessed more internal space for water than aggregates
consisting of rod-shaped {Fe(OH)3 } units and, consequently, it is more difficult to remove water from
Al-sww sludge than from Fe-sww. The results obtained showed that smaller particles dominate in
sww sludge, while larger particles are prevalent in nww sludge. To explain this distinct difference in
the size distribution of particles in sludge obtained with the use of Al3+ and Fe3+ , simple models of
aggregation and agglomeration-flocculation processes (aaf ) of treated wastewater have been proposed.
Except for PIX in nww, the analyzed particles of the investigated types of sludge were characterized by
similar specific surface area (Sps), regardless of the kind of sludge or the applied coagulant. Slightly
larger, negatively-charged sorg bridges, anchored directly on the surface of positive prtc are more
effective in closing the structure of nww sludge than small L bridges of the dye molecules anchored
on the surface of prtc via H2 PO4 − . All the discovered aspects could lead to improved performance of
wastewater treatment plants (WWTP) by increasing the efficiency of sludge dewatering.

Keywords: municipal wastewater sludge; dewatering; particle size distribution

Water 2019, 11, 101; doi:10.3390/w11010101 www.mdpi.com/journal/water

Water 2019, 11, 101 2 of 15

1. Introduction
Chemical coagulation is the second step in wastewater treatment [1,2]. This process has a
significant influence on the properties and structure of the resulting sludge [3–7]. The pH of sludge
decreases during chemical coagulation because Fe3+ or Al3+ undergoes cationic hydrolysis and its first
stage can be described as follows:

Fe3+ (Al3+ ) + HOH = FeOH2+ (AlOH2+ ) + H+ (1)

Positively-charged, colloidal particles {Fe(OH)3 } and/or {Al(OH)3 }, described further as prtc

(colloidal particles), are formed during the subsequent reactions which occur in wastewater [1,8].
These particles act as adsorbents [9] in the processes of aggregation and agglomeration-flocculation
(aaf ) of pollutants from wastewater. The surface charge of prtc is neutralized by negatively-charged
components such as H2 PO4 − , or by negatively-charged colloids, called sorg (organic substances). The
removal of neutral or positively-charged pollutants takes place during the so-called flocculation [1,5,10].
This mechanism dominates also in the case of wastewater treatment by electrocoagulation [11–13],
particularly during recirculating electrolysis of wastewater [14].
The quantity and quality of municipal wastewater sludge are indicators for the wastewater
treatment performance and, on the other hand, determine the possibilities for further utilization of
sludge [15]. For practical and technological reasons, the dewatering capability of sludge is extremely
important [16]. This property of municipal wastewater sludge very much depends on the type and
dose of inorganic coagulant added to the wastewater [17–19]. The structure of sludge flocs, as well
as their physical and chemical characteristics, determine the efficiency of the dewatering process of
municipal wastewater sludge [16,20]. Under laboratory conditions, so-called capillary suction time
(CST) is a measure of the sludge’s dewatering capacity [21]. Reproducibility and precision of CST
measurements [22,23] are important, both for theoretical considerations [24] and in practice, e.g., for
determination of an appropriate dose of an inorganic flocculant [25].
Probably the most important effect of flocs on the structure of municipal wastewater sludge
is through aggregation and agglomeration-flocculation (aaf ) processes [26–29]. Numerous papers
have been published, including information about measurements, modeling and characterization
of the structure of municipal wastewater sludge [30–33]. Often, the so-called fractal dimension D
becomes a specific research instrument [34,35]. For self-similar objects, e.g., sludge floc-aggregates
whose structure does not depend on a change in the scale, D is defined as follows:

M(R)~RD (2)

where M is the mass comprised in a sphere of the diameter R [34].

The value of D, either determined experimentally or calculated theoretically, has been used in
many theoretical considerations, e.g., in kinetic calculations [36], and also in practical solutions, e.g.,
for the filtration of excessive active sludge [37]. The structure and properties of municipal wastewater
sludge can also be examined directly through the determination of the distribution of the sizes of
particles and their specific surface [38–40].
Since it is suspected that there is an interpretable correlation between the structure of municipal
wastewater sludge and the value of CST corresponding to the sludge, this paper analyzes the observed
correlations and proposes simple models of aaf processes for chemically-coagulated synthetic and real
municipal wastewater.
The consecutive stages presented and considered in this article are based on:

(a) results of traditional jar tests,

(b) CST measurements,
(c) determination of the volumetric dimension (Dv) and respectively of the specific surface area (Sps).
Water 2019, 11, x FOR PEER REVIEW 3 of 15
Water 2019, 11, 101 3 of 15
2. Materials and Methods
Synthetic
2. Materials anddyeing wastewater (sww) and real municipal wastewater (nww) originating from a
Methods
wastewater treatment plant in Reszel (North-Eastern Poland) were investigated. The municipal
Synthetic dyeing wastewater (sww) and real municipal wastewater (nww) originating from a
wastewater had the following parameters: suspended solids: SS (mg/L) = 250–800; total phosphorus:
wastewater treatment plant in Reszel (North-Eastern Poland) were investigated. The municipal
P (mg/L) = 9–13.5, chemical oxygen demand: COD (mgO2/L) = 600–1800; turbidity: TU = 80–160 NTU;
wastewater had the following parameters: suspended solids: SS (mg/L) = 250–800; total phosphorus:
and pH = 6.6–7.8. In turn, each 1 L of sww contained 31.3 mg H2PO4− (10 mg P) and 50 mg disperse
P (mg/L) = 9–13.5, chemical oxygen demand: COD (mgO2 /L) = 600–1800; turbidity: TU = 80–160 NTU;
dye (L) (Synthene Scarlet P3GL) produced by the Boruta-Zachem Chemical Company, from Zgierz
and pH = 6.6–7.8. In turn, each 1 L of sww contained 31.3 mg H2 PO4 − (10 mg P) and 50 mg disperse dye
(Poland). Therefore, the composition of synthetic dyeing wastewater (31.3 mg H2PO4− + 50 mg
(L) (Synthene Scarlet P3GL) produced by the Boruta-Zachem Chemical Company, from Zgierz (Poland).
disperse dye/L) is a result of the synergy effect of both components; a mixture of dye and municipal
Therefore, the composition of synthetic dyeing wastewater (31.3 mg H2 PO4 − + 50 mg disperse dye/L)
wastewaters is susceptible on coagulation and/or electrocoagulation, which has been proved and
is a result of the synergy effect of both components; a mixture of dye and municipal wastewaters
explained by previous studies [41]. It was experimentally demonstrated that in the absence of P-PO4
is susceptible on coagulation and/or electrocoagulation, which has been proved and explained by
it is impossible to remove even a small amount of dye L by chemical coagulation of its aqueous
previous studies [41]. It was experimentally demonstrated that in the absence of P-PO4 it is impossible
solution. However, coagulation of L solution proceeds effectively and efficiently in the presence of
to remove even a small amount of dye L by chemical coagulation of its aqueous solution. However,
phosphate ions [42] and for this reason we decided to use this composition of synthetic wastewater
coagulation of L solution proceeds effectively and efficiently in the presence of phosphate ions [42]
in the present study.
and for this reason we decided to use this composition of synthetic wastewater in the present study.
The sludge samples submitted for further tests were obtained in wastewater coagulation by: (a)
The sludge samples submitted for further tests were obtained in wastewater coagulation
PIX 113, a ferric coagulant (based on Fe2(SO4)3) widely used in Poland, and (b) PAX 18, pre-
by: (a) PIX 113, a ferric coagulant (based on Fe2 (SO4 )3 ) widely used in Poland, and (b) PAX
polymerised aluminum coagulant, an alternative to PIX. Both coagulants were produced by Kemipol,
18, pre-polymerised aluminum coagulant, an alternative to PIX. Both coagulants were produced
the Polish branch of the Kemira Chemicals in Gdansk, Poland. By adding the above-mentioned
by Kemipol, the Polish branch of the Kemira Chemicals in Gdansk, Poland. By adding the
coagulants to the wastewaters, many intermediate polymeric species such as (Al(OH)3)n, (Fe(OH)3)m
above-mentioned coagulants to the wastewaters, many intermediate polymeric species such as
or (Fe(OH)2)p are produced, being responsible for colloidal sorption on their surfaces of pollutants
(Al(OH)3 )n , (Fe(OH)3 )m or (Fe(OH)2 )p are produced, being responsible for colloidal sorption on
from wastewaters.
their surfaces of pollutants from wastewaters.
In sww, the dye concentration was determined spectrophotometrically at a wavelength λ = 460
In sww, the dye concentration was determined spectrophotometrically at a wavelength λ = 460 nm,
nm, while P-PO4 was assayed according to the standard method (λ = 690 nm) using a HACH DR 3900
while P-PO4 was assayed according to the standard method (λ = 690 nm) using a HACH DR 3900
instrument (Hach Company, Loveland, CO, USA) with 13 mm standard cell tests.
instrument (Hach Company, Loveland, CO, USA) with 13 mm standard cell tests.
After 30 min of sedimentation, followed by decantation, 25.0 cm33 of separated sludge was
After 30 min of sedimentation, followed by decantation, 25.0 cm of separated sludge was
collected in order to determine capillary suction time CST [21–23], which is a measurement of the
collected in order to determine capillary suction time CST [21–23], which is a measurement of the
dewaterability of sludge. In CST measurements, a new prototype developed in the University of
dewaterability of sludge. In CST measurements, a new prototype developed in the University of
Warmia and Mazury in Olsztyn, Poland (DWTEST—Dewatering tester, schematically illustrated in
Warmia and Mazury in Olsztyn, Poland (DWTEST—Dewatering tester, schematically illustrated in
Figure 1), was used.
Figure 1), was used.

Figure 1. The schematic diagrame of a new DWTEST prototype for measurement of the capillary
Figure time
suction 1. The schematic diagrame of a new DWTEST prototype for measurement of the capillary
(CST).
suction time (CST).
Water 2019, 11, 101 4 of 15

Water 2019, 11, x FOR PEER REVIEW 4 of 15

In each case, 25 cm3
of separated sludge was applied onto a Whatman filter paper disc, while
In each
pressing case, 25 cm3button
the appropriate of separated sludge wasAfter
on the apparatus. applied ontodozen
several a Whatman
seconds,filter paperofdisc,
the value while
measured
pressing the appropriate
time corresponding button
to the flow ofonthe
the liquid
apparatus. Afterelectrodes
between several dozen
in theseconds,
measuringthe cell
value of measured
was displayed.
time corresponding to the flow of the liquid between electrodes in the
The measurements were always repeated three times and the values for standard deviation measuring cell was displayed.
(SD) (s)
The measurements
are indicated in thewere always
graphs. repeated
The results of three times and the values
CST measurements for standard
are presented in the deviation
graphical (SD)
form(s)
in
are indicated in the graphs. The results
dependence of coagulant dose CST = f(mg Al or Fe/L). of CST measurements are presented in the graphical form in
dependence
Particleof coagulant
size doseofCST
distribution the =sludge
f(mg Al wasordetermined
Fe/L). by measurement of laser light dispersion
Particle size distribution of the sludge was determined
using a Mastersizer 3000 unit, Malvern Instruments, Malvern, United by measurement
Kingdom of laser
[43]. light dispersion
Sludge samples
using
were instilled to a measuring cell until an obscuration of 5–15% was achieved. The refractivesamples
a Mastersizer 3000 unit, Malvern Instruments, Malvern, United Kingdom [43]. Sludge indices
were instilled
for water and to a measuring
wastewater were cell1.33
until
and an3.80,
obscuration of 5–15%
respectively. was achieved.
The particle The refractive
size distribution was indices
used to
for water and
determine thewastewater
mean particle weresize1.33 and
D(3,2) 3.80,and
(µm) respectively. The particle
available surface size distribution
area (Sps) of particles inwastheused to
sludge
determine
2 the mean particle size D (3,2) (μm) and available surface area (Sps) of particles in the sludge
(m /g). Mean particle size is defined as [43]:
(m2/g). Mean particle size is defined as [43]:
D(3,2) =n
D(3,2) diid3 i/Σn
= in 3/Σniid
2
dii2 (3)
(3)
where nii is
is the
the number
number of
of particles
particles of
of diameter
diameter dii..
Volumetric
Volumetric dimensions Dv10, Dv50 and
dimensions Dv10, Dv50 and Dv90
Dv90 denoting
denoting the
the maximum
maximum particle
particle diameter
diameter below
below
which
which particles
particles account
account for
for 10%,
10%, 50%
50% and
and 90%
90% of of the
the volume
volume ofof the
the analyzed
analyzed sludge,
sludge,respectively,
respectively,
were
were also calculated.
calculated. The
The results
results represent
represent the
the mean
mean values
values of
of three
three replications
replications for
for each
each type
type of
of
sludge
sludge evaluated.
evaluated.

3. Results
3. Results and
and Discussion
Discussion
The results
The results ofof the laboratory
laboratory tests are presented in Tables 11 and
and 22 and
and some
some of
of them
themininFigure
Figures2.
The
2. mean
The meanvalues
valuesfrom
fromthree repetitions
three were
repetitions wereused forfor
used plotting the
plotting graphs
the graphsand SD
and SD values (in(in
values %%oror
s)
were marked in each case.
s) were marked in each case.
Figure 22 illustrates
Figure illustrates the effects of coagulation of sww and nww by coagulants PIX and PAX. PAX.

sww PIX nww PIX

Removal Removal
[%] [%]
100 100

80 D 80 SS
60 60

40 40

20 2,38 20 3,02
P mg Fe/L P ↑ mg Fe/L
0 ↑ 0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6

sww PAX nww PAX

Removal Removal
[%] [%]
100 100

80
D 80 SS
60 60

40 40
20 1,84 20
P mg Al/L P mg Al/L
0 ↑ 0 1,61
0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5↑ 2 2.5

Figure 2. Synthetic wastewater (sww) and real municipal wastewater (nww) coagulated by PIX and PAX.
Figure 2. Synthetic wastewater (sww) and real municipal wastewater (nww) coagulated by PIX and
PAX.
The initial (o ) and final (f ) values of the pH, concentration of total phosphorus (P), chemical
oxygen demand (COD) and suspended solids (SS) are presented for both PIX and PAX coagulants
usedThe initial (o) of
for treatment and final
both sww(f) and
values
nww:of the pH, concentration of total phosphorus (P), chemical
oxygen demand (COD) and suspended solids (SS) are presented for both PIX and PAX coagulants
used for treatment of both sww and nww:
Water 2019, 11, 101 5 of 15

PIX-sww: pHo = 5.2 → pHf = 4.75 PAX-sww: pHo = 5.2 → pHf = 5.0

PIX-nww: SSo = 750, Po = 11.40; CODo = 1690 → CODf = 360 (mgO2 /L);

pHo = 7.53 → pHf = 6.84

PAX-nww: SSo = 440, Po = 9.75; CODo = 1190 → CODf = 313 (mgO2 /L);

pHo = 7.80 → pHf = 7.14

Courses of all four dependences, where Removal = f(coagulant dose), showed that the highest
dose of both aluminum ions and ferric ions ensured 100% removal of L from sww and SS from nww.
Under these conditions, PIX removed 86–87%, and PAX 96–99% of phosphorus (P) from both sww and
nww. Because the limit of P allowed in the effluents according to Polish legislation for wastewater
treatment plants (WTTP) is 5 mg P/L, the sludge analyzed in our research was precipitated under
conditions ensuring 50% removal of P from treated wastewater. It can be assumed that primary
coagulation of municipal wastewater in Poland is sufficient to obtain the required P level in effluent,
principally owing to the mentioned 50% coagulation of phosphorus compounds. In all of the four
diagrams seen in Figure 2, an arrow on the x-axis indicates the numerical value of the dose of a
coagulant (fes, fen, als, aln) adequate for attaining 50% of phosphorus removal from sww and nww.
When these doses are added, i.e., to sww: fes = 2.38 mg Fe/L ≡ 0.0425 mmol Fe/L and als =1.84 mg
Al/L ≡ 0.068 mmol Al/L; and to nww: fen = 3.02 mg Fe/L ≡ 0.054 mmol Fe/L and aln = 1.61 mg Al/L
≡ 0.060 mmol Al/L, sediments were obtained, which were subsequently tested to determine CST, Sps
and Dv (Table 1) as well as percentage shares of particular particle sizes.
During the coagulation process of sww, the presence of phosphate ions in the system made it
possible to effectively remove the dye from the liquid phase of wastewater. Phosphate anions are
adsorbed on positively-charged colloidal particles {Fe(OH)3 } and/or {Al(OH)3 } (prtc), creating units of
the type:
{Fe(OH)3 } − (H2 PO4 − )c and {Al(OH)3 } − (H2 PO4 − )d (4)

As the negative sorbate (phosphate) accumulates on prtc, the positive potential of the systems in
Formulas (4) decreases and mutual repulsive forces weaken between particles, which determine the
direction, range and intensity of Brownian motions. Thus, the stability of the colloidal system decreases,
while the probability of collisions between particles rises and finally an aaf process may occur.
In practice, an aqueous solution of phosphate ions undergoes coagulation (both precipitation and
adsorption) and almost all phosphates could be transferred into municipal wastewater sludge.
As mentioned at the beginning of the Materials and Methods Section, in the absence of P-PO4 it
was impossible to remove even a small amount of a dye directly, by chemical coagulation of a water
solution of L, whereas the coagulation or electro-coagulation of L [42] proceeded efficiently only when
“supported” by phosphate ions. To explain the reasons for this phenomenon, the following simple
model of adsorption of a dye particle to systems is proposed in Formulas (5) and (6):

{Fe(OH)3 } − (H2 PO4 − )c − L (5)

{Al(OH)3 } − (H2 PO4 − )d − L (6)

This model assumes a slightly positive surface charge of a dye molecule, which repels it from a
positive prtc while attracting it to prtc centers absorbing H2 PO4 − . It is thus based on the assumption that
dye molecules can be adsorbed to units, Formulas (4), by forming bridges with the help of previously
adsorbed H2 PO4 − ions (Formulas (5) and (6)). Although the occurrence of dye molecules, bridged
on the surface of prtc with the help of H2 PO4 − slightly decreases the process of neutralization of the
system’s surface, it also causes the growth of its mass and size, which is conducive to destabilization
of prtc and leads to aaf. Thus, in laboratory practice, slightly more coagulant is used to achieve
destabilization of a mixture of dye and H2 PO4 − than for a solution containing only H2 PO4 − .
Water 2019, 11, 101 6 of 15

Figure 3a shows volumetric proportions for a spherical colloidal particle {Al(OH)3 } (sized from
86 to 206 µm, for which the diameter was assumed to be around 165 nm [44], which in fact, has a
Water 2019, 11, x FOR PEER REVIEW 6 of 15
rather corrugated surface), a molecule of the dye (having an estimated 1 mole ≈ 500 g and a size of
about 10 nm) andanda Ha2HPO −
about 10 nm) 4 4−ion
2PO ion(with
(withaadiameter
diameter of of 1.2
1.2 nm). In the
nm). In the subsequent
subsequentmodels
modelspresented
presentedinin
thisthis
paper (Figure 3b), for better clarity of the visualization, rod-shaped particles {Fe(OH)
paper (Figure 3b), for better clarity of the visualization, rod-shaped particles {Fe(OH)3}3[45] } [45]and
and
spherical particles
spherical {Al(OH)
particles 3 } [44]
{Al(OH) 3} are
[44]presented in appropriately
are presented diminished
in appropriately sizes, which
diminished is highlighted
sizes, which is
by using a broken line to draw their contours.
highlighted by using a broken line to draw their contours.
As As
expected,
expected, amounts
amounts ofofmg mgFe/L
Fe/Lfrom
fromPIX
PIX (fes and fen) needed
(fes and needed totoremove
remove50%50%ofofPPwere
werehigher
higher
than thethe
than respective
respective amounts
amounts ofofmgmgAl/L
Al/Lfrom
fromPAX
PAX (als and aln). Simultaneously,
(als and Simultaneously,the thesame
sameamounts
amounts
expressed
expressed in mmol/Lwere
in mmol/L werehigher
higherfor forPAX
PAXthan
than for
for PIX.
PIX. In Poland
Poland PIX,
PIX,rather
ratherthan
thanPAX,
PAX,isisa amore
more
popular coagulant in urban wastewater treatment plants (despite being less efficient) because ofitsits
popular coagulant in urban wastewater treatment plants (despite being less efficient) because of
muchmuch lower
lower price
price perper 1 kg
1 kg ofof coagulant.
coagulant.

(a) (b)
Figure
Figure 3. Schematic
3. Schematic diagramsofofinteractions
diagrams interactions between
between colloidal
colloidal particles
particles and
andaadye
dyemolecule
moleculeandand
H 2 −
PO 4− ion, respectively: (a) Approximate proportions between sizes of: colloidal particle {Al(OH)3},
H2 PO4 ion, respectively: (a) Approximate proportions between sizes of: colloidal particle {Al(OH)3 },
dyedye L molecule
L molecule andand
H2H PO2PO−4− ion; (b) schemes of structures for Formulas (4) and (5) responsible for
4 ion; (b) schemes of structures for Formulas (4) and (5) responsible for
adsorption of H 2PO4 and L from sww on the surface of colloidal particles (prtc).
− −
adsorption of H 2 PO4 and L from sww on the surface of colloidal particles (prtc).

Slightly
Slightly moremore mg/LPIX
mg/L PIXwaswasapplied
appliedto to remove
remove 50% 50% P P from
fromnww nwwthan
thanfrom
fromsww.
sww.Most
Most probably,
probably,
a part
a part of of
thethePIX
PIXdose
dose in
in nww
nwwhad hadbeen
been sacrificed
sacrificed to remove
to remove1.3301.330
mg/Lmg/L
sorg denoted as COD;asCOD
sorg denoted COD;o

= 1690 → COD f = 360 mgO2/L (Figure 2). Meanwhile, the same sample was characterized by a
CODo = 1690 → CODf = 360 mgO2 /L (Figure 2). Meanwhile, the same sample was characterized by a
slightly
slightly higher
higher concentrationofofphosphorus
concentration phosphorus (P
o 11.4→
(Po ==11.4 → PPff==0.64
0.64mg/L)
mg/L) and SSoSS
and (750(750
o
mg/L) than the
mg/L) than
sample treated by PAX (P o = 9.75 → Pf = 0.22 mg/L, SSo = 440 mg/L). For nww a slightly lower dose
the sample treated by PAX (Po = 9.75 → Pf = 0.22 mg/L, SSo = 440 mg/L). For nww a slightly lower
of PAX in mg/L, than from sww, was needed to remove 50% of phosphorus, which was most probably
dose of PAX in mg/L, than from sww, was needed to remove 50% of phosphorus, which was most
a consequence of lower consumption of the coagulant for removing just 887 mgO2/L COD from nww
probably a consequence of lower consumption of the coagulant for removing just 887 mgO2 /L COD
(CODo = 1190 → CODf = 313 mgO2/L) at an approximately similar efficiency as PIX in the removal
from nww (CODo = 1190 → CODf = 313 mgO2 /L) at an approximately similar efficiency as PIX in the
of L from sww.
removal of L from sww.
Figure 4 contains a schematic representation of the removal of phosphorus as well as sorg from
Figure
nww using 4 contains a schematic
PIX and PAX, representation
respectively. Unlike L beingof the removal
bridged of phosphorus
in sww sorg from
as well as charged
by H2PO4−, negatively
nww using PIXare
anddirectly
PAX, respectively. in sww by −
sorg in nww adsorbed on Unlike L being
the surface bridged
of prtc. The schemas H2 PO
shown , negatively
in4Figure charged
4 also account
sorgfor nww
in the are directly adsorbed on the surface of prtc.
hydrophilic properties of wastewater sol (sorg), marking in yellow the water molecules account
The schemas shown in Figure 4 also which
for stabilize
the hydrophilic properties
this sol. Because of wastewater
of the sol (sorg),
negative charge of sorgmarking
here (nww),in yellow
fewer Hthe
2PO water moleculesonwhich
4− are adsorbed the
stabilize this of sorg − are adsorbed on the
surface of sol.
prtc Because
(only 7 Hof2POthe4− negative charge
ions in the schema here (nww),
in Figure fewer H
4), compared PO4respective
to2the surfaces of
surface of prtc (only 7H −
prtc formed during 2 PO
sww 4 ions in the schema in Figure 4), compared to the respective surfaces of
coagulation.
prtc formed during sww coagulation.
Water 2019, 11, 101 7 of 15
Water 2019, 11, x FOR PEER REVIEW 7 of 15

sorg

{Al(OH)33} sorg

{Fe(OH)33}
sorg

Figure 4. Schemas
Schemas ofof structures
structures responsible
responsible for
foradsorption
adsorptionof 2PO44−
ofHH22PO 4 andand organic
organic substances
substances (sorg)
−−
Figure 4. (sorg)
from nww
nww onon the
the surface
surface of
of prtc.
prtc.

The following
following graphs
graphsininFigure
Figure5 5show
showthe
theprocess of of
process dewatering sww
dewatering nww,
andand
sww presented
nww, in the
presented in
form
the of CST
form = f(dose
of CST of coagulant)
= f(dose relationships.
of coagulant) relationships.

sww PIX nww PIX

CST [s] yy == 4.3284x
4.3284x ++ 43.5
43.5 CST [s] yy == -20.612x
-20.612x ++ 139.25
139.25
R²
R² == 0.9917
0.9917 R²
R² == 0.9899
0.9899
140
140 140
140
120
120 120
120
100
100 100
100
80
80 80
80
60
60 60
60
40
40 40
40
20 2.38
2.38 3.02
3.02
20 mg Fe/L 20
20 mg Fe/L
00 ↑
↑ 00 ↑
↑
00 11 22 33 44 00 11 22 33 44

sww PAX nww PAX

CST [s] yy == 19.217x
19.217x ++ 14.32
14.32 CST [s] yy == -34.283x
-34.283x ++ 135.27
135.27
R²
R² == 0.9733
0.9733 R²
R² == 0.9697
0.9697
140
140 140
140
120
120 120
120
100
100 100
100
80
80 80
80
60
60 60
60
40
40 40
40
20
20 1.84
1.84 20 1.61
1.61
mg Al/L 20 mg Al/L
00 ↑
↑ 00 ↑
↑
00 0.5
0.5 11 1.5
1.5 22 2.5
2.5 00 0.5
0.5 11 1.5
1.5 22 2.5
2.5

Figure 5. Dewatering of sludge obtained from sww and nww coagulated with PIX and PAX.
Figure 5. Dewatering of sludge obtained from sww and nww coagulated with PIX and PAX.
In both cases, the doses of coagulants (0.67 to 3.35 mg Fe/L and 0.46 to 2.3 mg Al/L) are within
In both
the range cases, used
of doses the doses
in theofcoagulation
coagulantstests (0.67illustrated
to 3.35 mginFe/L Figure and2.0.46 to 2.3 mg
According Al/L) are
to Figure within
5, the low
the range of doses used in the coagulation tests illustrated in Figure 2. According
values of SD (s), show a high repeatability of the measurements for capillary suction time CST, carried to Figure 5, the low
values of SD (s), show
out by DWTEST a high repeatability
instrument. of the measurements
For sww wastewater, the tendency for capillary
of the suctionCST
relationship time CST, carried
sww = f(dose of
out by DWTEST
coagulant) instrument.
demonstrates For increase
a linear sww wastewater,
in CST, when the tendency
the dose of of coagulant
the relationship CSTsww
is increased. = f(dose of
swwConversely,

coagulant) demonstrates a linear increase in CST, when the dose

the increase in the dose of added coagulate to nww causes a linear decrease in CSTnww . The high of coagulant is increased.
values
Conversely, the increase in the dose of added
2 coagulate to nww causes
of the regression correlation coeficient R , (which are ranged between 0.9697 and 0.9917), provide a linear decrease in CSTnwwthe.
nww

The
linearhigh valuesconfirmation
character of the regression correlation
for the regressioncoeficient
relationship R , (which
2
2
CST = f(doseare ranged between 0.9697
of coagulant). For fes,andthe
0.9917), provide the linear character
CSTsww was 53.5 s, and for als it was 58 s. confirmation for the regression relationship CST = f(dose of
coagulant). For fes, the
With increasing CSTsww
doses of was 53.5 s,CST
swwcoagulant, andsww for increased
als it was 58 from s. 46 to 58 s for PIX and over a slightly
With increasing doses of coagulant, CST
wider range from 25 to 61 s for PAX. Undoubtedly, the structure
sww
sww increased from 46 of
to sww
58 s for
sludgePIX and over awith
obtained slightly
PIX
wider range from 25 to 61 s for PAX. Undoubtedly, the structure of
is different from the structure of sludge achieved with PAX, although the mechanism of bridging sww sludge obtained with PIX is
different from
particular unitsthe structure
is similar. of sludge
As the dose ofachieved
a coagulant with PAX, although
increased, the share theof mechanism
these units inofstructures
bridging
particular units is similar. As the dose of a coagulant increased, the
of sludge increased, which may have led to the blocking of water molecules in sludge flocs and ashare of these units in structures
of sludge increased,
subsequent increase which
in CST.may It is have
known ledthatto the blocking
micelles of water
{Fe(OH) molecules in sludge flocs and a
3 }n and {Al(OH)3 }n differ from each
subsequent
other in shapeincrease in CST. It is[44,45].
and dimension knownThe thatregression
micelles {Fe(OH)
equations 33}nn and {Al(OH)
(Figure 33}nn differ
5) allow fromtoeach
the CST other
be easily
in shape and dimension [44,45]. The regression equations (Figure 5) allow
calculated for an identical dose of both coagulants, which is 0.05 mmol/L (within the range of fes = 0.045 the CST to be easily
calculated for an identical dose of both coagulants, which is 0.05 mmol/L (within the range of fes =
0.045 and als = 0.068): CSTFe Fe = 55.5 s, and CSTAl Al = 40.3 s. The next calculation can be made for the same
Water 2019, 11, 101 8 of 15

Water 2019, 11, x FOR PEER REVIEW 8 of 15

and als = 0.068): CSTFe = 55.5 s, and CSTAl = 40.3 s. The next calculation can be made for the same
dose
doseofofboth
both coagulants,
coagulants, equal
equal to
to 0.1 mmol/L. Here,
0.1 mmol/L. Here, CST
CSTFe = 67.7 s, and CSTAl = 66.2 s, which almost
Fe = 67.7 s, and CSTAl = 66.2 s, which almost
the
thesame.
same.ByByextrapolating
extrapolatingthetheCST
CSTvalues
valuestotohigher
highercoagulant
coagulantdoses,
doses,ititcan
canbe
behypothesized
hypothesizedthat
thatsww
sww
sludge
sludge with PIX binds water more effectively at lower Fe doses, while sww sludge with PAXbinds
with PIX binds water more effectively at lower Fe doses, while sww sludge with PAX binds
water
watermore
moreeffectively
effectivelyat
athigher
higherAlAldoses.
doses.Due
Duetotothe
thesmall
smallnumber
numberof ofcomponents
componentsininsww
swwsludge
sludgeasas
described in Formula (7):
described in Formula (7):
water
water+ +dye
dye++KH
KH2 PO
2PO + prtc
44 + prtc (7)
(7)
theonly
the onlyexplanation
explanationof ofthetheCST
CST==f(dose
f(doseof ofcoagulant)
coagulant)isisthe thestructure
structureof ofthis
thissludge.
sludge.
Thestructures
The structuresillustrated
illustratedininFigureFigure6 6schematically
schematicallydescribe
describethe the progressing
progressing aafaaf processes.
processes. It isIt
is clear
clear thatthat repulsion,
repulsion, as well as well
as theasintensity
the intensity
and scopeand of scope of Brownian
Brownian motions, motions,
decreasedecrease
as the doseas theof
dose of a coagulant increases. The network of bridging and connections
a coagulant increases. The network of bridging and connections grows in the emerging aggregates- grows in the emerging
aggregates-flocs
flocs of sedimentofsww. sediment sww. The
The schemas inschemas
Figure 6inmay Figure 6 may the
illustrate illustrate the destabilization
destabilization of a singleof acluster
single
cluster (previously
(previously presentedpresented
in Figure in Figure
3) due3)to duetheto formation
the formation of appropriate
of appropriate dimers,
dimers, in inwhich
whicha adyedye
moleculemost mostprobably
probably can can bridge
bridge two prtc partly destabilized by −
molecule by HH22PO4 . .At
− Atananincreasing
increasingdose doseof of
aacoagulant,
coagulant,thesethesestructures
structuresare aremost
mostprobably
probably bridging,
bridging, thus
thus binding
binding water
water in the
in the internal
internal spaces
spaces of
of agglomerate-flocs.
agglomerate-flocs. It isIt known
is known that
that rod-shaped
rod-shaped colloidalparticles,
colloidal particles,e.g.,
e.g.,{Fe(OH)
{Fe(OH)3},3 },coagulate
coagulatemore more
rapidlythan
rapidly thanspherical
sphericalunitsunits{Al(OH)
{Al(OH)33}.}.However,
However,the theaggregation
aggregationof ofmany
many“rods” “rods”ultimately
ultimatelyleads
leadsto to
the formation ofofa aspherical
the spherical aggregate-floc.
aggregate-floc. MostMostprobably, spherical
probably, aggregates
spherical composed
aggregates of spherical
composed of
units {Al(OH)
spherical units3 }{Al(OH)
leave more internal
3} leave morespace for water
internal spacethanfor other
water spherical
than other aggregates
sphericalcomposed
aggregates of
rod-shaped
composed of{Fe(OH)
rod-shaped 3 }, which
{Fe(OH)is why it is more
3}, which is why difficult to remove
it is more difficultwater from Al-sww
to remove water from(longer CST)
Al-sww
than from
(longer CST) Fe-sww (shorter
than from Fe-swwCST). (shorter CST).

Schemaofofthe
Figure6.6.Schema
Figure theprogress
progressof aafsww.
ofaaf sww.

Asmentioned
As mentionedabove,above,in incontrast
contrastto swwsludge,
tosww sludge,an anincrease
increasein inthe
thedose
doseofofaacoagulant
coagulant(both(bothPIX
PIX
and PAX) added to nww caused a linear decrease CSTnww. The CSTsww for fen was 77 s, and for alnaln
and PAX) added to nww caused a linear decrease CST nww . The CST sww for fen was 77 s, and for it
it was
was 80 80 s. the
s. As As thedosedose
of aof a coagulant
coagulant increased,
increased, the CST
the CST nww decreased
nww decreased from from
124 to124
70 to 70 PIX,
s for s for and
PIX,
and from
from 115 to115
54 stofor
54PAX.
s for In
PAX. In comparison
comparison with sww,withinsww,
this in
casethis
thecase the variation
variation of CSTfor
of CST values values
sludgefor
sludge obtained with PIX and PAX was distinctly smaller. From an appropriate
obtained with PIX and PAX was distinctly smaller. From an appropriate regression equation (Figure regression equation
(Figure
5), 5), for identical
for identical doses
doses of both of both coagulants
coagulants (0.05 the
(0.05 mmol/L), mmol/L),
following thevalues
following values
of CST wereofcalculated:
CST were
calculated: CST = 81.5 s and CST = 89 s. Analogously to sww tests, the
CSTFe = 81.5 s and CSTAl = 89 s. Analogously to sww tests, the subsequent calculations were made for
Fe Al subsequent calculations
were
the made
same doseforofthe
thesame dose ofequal
coagulants the coagulants
to 0.1 mmol/L.equalTheto 0.1 mmol/L.
results were TheCSTresults were CSTFeAl==23.8
Fe = 23.8 s and CST 42.7s
s,and CSTmeans
which Al = 42.7thats, CST
which means that CSTAlthan
Al is distinctly higher
is distinctly
CSTFe. Whenhigher than CSTFe .CST
extrapolating When extrapolating
values to higher
coagulant doses, it appears that at higher coagulant doses nww sludge with PAX may sludge
CST values to higher coagulant doses, it appears that at higher coagulant doses nww with
bind water
PAX may bind water more effectively than nww sludge with PIX. Compared
more effectively than nww sludge with PIX. Compared to sww, nww sludge is a much more complex, to sww, nww sludge
is a much more complex,
multi-component system. Frommulti-component
this point of view,system.the From this point constituent
most significant of view, theofmost
nwwsignificant
sludge is
sorg. Most probably, the participation of sorg in the aaf process, leading to the formation of nww
Water 2019, 11, 101 9 of 15

Water 2019, 11, x FOR PEER REVIEW 9 of 15

constituent of nww sludge is sorg. Most probably, the participation of sorg in the aaf process, leading to
sludge, is responsible
the formation for theis negative
of nww sludge, responsibleregression of the course
for the negative regressionof the CST
of the = f(dose
course of theofCST
coagulant)
= f(dose
function.
of coagulant) function.
Figure
Figure 77 shows
shows aa suggested
suggested schemaschema of of the
the aaf processes
processes occurring
occurring during
during nww nww coagulation.
coagulation.
Similarly
Similarly to to sww,
sww, anan increase
increase in in the
the dose
dose ofof aa coagulant
coagulant leads leads to to aa decrease
decrease in in repulsion
repulsion as as well
well as
as the
the
intensity and scope of Brownian motions in the system. The network of
intensity and scope of Brownian motions in the system. The network of branches and connections in the branches and connections in
the
swwsww sludge
sludge is expanding.
is expanding. TheThe schema
schema in Figure
in Figure 7 may7 may illustrate
illustrate the destabilization
the destabilization of a single
of a single unit
unit from
from
Figure 4 through the formation of appropriate dimers, in which sorg most probably bridges two two
Figure 4 through the formation of appropriate dimers, in which sorg most probably bridges prtc
prtc partly destabilized
by H2by POH − At4−an
partly destabilized 4 2.PO . At an increasing
increasing dose ofdose of the coagulant,
the coagulant, these structures
these structures most
most probably
probably begin to
begin to branch, branch,
binding binding
water water spaces
in internal in internal
within spaces within agglomerates-flocs.
agglomerates-flocs. Organic
Organic compounds,
compounds,
sorg, which in total constitute so-called the “negative wastewater colloid” of nww, are classifiedare
sorg, which in total constitute so-called the “negative wastewater colloid” of nww, as
classified
hydrophilic as hydrophilic colloids,
colloids, stabilized bystabilized
the hydrationby the hydration
shell of water shell of waterAs
molecules. molecules.
the negativeAs the negative
wastewater
wastewater colloid is destabilized,
colloid is progressively progressivelyand destabilized, and the
the aaf processes are inaafprogress,
processes theare in progress,
concentration the
of sorg
concentration of sorg decreases and water molecules released from
decreases and water molecules released from the hydration shell become “available”, which leads the hydration shell become
“available”,
to a decreasewhich in CST.leads
Theto a decrease
increasing in CST. The
difference increasing
in the values ofdifference
CSTFe andinCST theAlvalues of CSTFedoses
at increasing and
CST at increasing doses of a coagulant can be explained in this case as
of a coagulant can be explained in this case as for sww. Gradually branching (Figure 7) structures
Al for sww. Gradually branching
(Figures
containing 7) spherical
structures{Al(OH)
containing spherical {Al(OH)3} units absorbing components of nww block the
3 } units absorbing components of nww block the “inner-network” water
“inner-network”
more effectively than water more effectively
rod-shaped than{Fe(OH)
units with rod-shaped3 }.
units with {Fe(OH)3}.

Figure 7.
Figure Schema of
7. Schema of the
the progress
progress of aaf processes
of aaf processes in nww.
in nww.

Figure 88 shows
Figure shows thethe percentages
percentages of of specific
specific size
size classes
classes of
of particles
particles in
in sludge
sludge obtained
obtained in in the
the
coagulation of sww and nww; using a) fes and als, and b) fen and aln. For both coagulants
coagulation of sww and nww; using a) fes and als, and b) fen and aln. For both coagulants and in both and in both
types of
types of wastewater,
wastewater, twotwo classes
classesof
ofsludge
sludgeparticle
particlesize
sizecan
canbebedistinguished.
distinguished.The Thesww
swwtypetypeofof
sludge
sludge is
dominated by particles of a smaller size: 3–40 µm, but with a maximum of 3.72%
is dominated by particles of a smaller size: 3–40 μm, but with a maximum of 3.72% at 12.7 μm for at 12.7 µm for PIX,
and 5–75
PIX, and µm5–75with
μm awith
maximum of 6.38%
a maximum of at6.38%
24.1 µm for PAX.
at 24.1 On PAX.
μm for the other
On hand, nwwhand,
the other is dominated
nww is
dominated by sludge particles larger in size, in the range of 45–1300 μm for both coagulants. For low
by sludge particles larger in size, in the range of 45–1300 µm for both coagulants. For the very the
doseslow
very of adoses
coagulant
of a applied,
coagulant theapplied,
dominant thesize of nww size
dominant sludge particles
of nww within
sludge the first
particles rangethe
within of sizes
first
is difficult
range to define.
of sizes For to
is difficult both coagulants,
define. For bothabout 3.5% ofabout
coagulants, sludge particles
3.5% are sized
of sludge about
particles are130 µm,
sized and
about
are located within the second range.
130 μm, and are located within the second range.
Distributions of sludge particle size classes were very similar to the ones achieved for nww which
were obtained in real municipal wastewater coagulated with the minimal doses of PIX (13.55 mg Fe/L)
and PAX (2.45 mg Al/L), respectively [40]. It can be observed that these doses were much higher than
the doses used in this study (3.02 mg Fe/L and 1.61 mg Al/L, respectively).
At the lower coagulant dose, larger size particles > 100 µm were prevalent as well. On the other
hand, distributions of sludge particle classes similar to the ones identified in sww have also been
recorded previously [39] in sludge obtained from synthetic wastewater coagulated with minimal doses
Water 2019, 11, 101 10 of 15

of PIX (6.5% of particles about 4 µm in size) and PAX (10% of particles about 10 µm in size). Certain
similarities and differences in particle size classes between sww and nww sludge discussed here are
most probably a consequence of very low doses of coagulant; in this study, a coagulant dose was high
enough to remove
Water 2019, 11, x FORonly
PEER50% of P. Development of this issue will be completed by using the values
REVIEW 10 Sps
of 15
and Dv, which are collected in Table 1.

sww nww

7
7
PAX
6
6
5
5
volume [%]

volume [%]
4 PAX
4

3 PIX 3

2 2 PIX

1 1

0 0
0.01 1 100 10000 0.01 1 100 10000
Particle size classes [μm] Particle size classes [μm]

(a) (b)
Figure
Figure 8. 8.Sizes
Sizesofofflocs
flocsofofsludge
sludgeobtained
obtained from
from sww
sww and
and nww
nww coagulated
coagulated with
with the
the help
help of:of:(a)(a)
fesfes and
and
als,
als, (b)
(b) fenfen and
and aln.
aln.

Table 1. The specific surface area (Sps) and volumetric dimension (Dv) values, for sludge particles
Distributions of sludge particle size classes were very similar to the ones achieved for nww which
illustrated in Figure 8.
were obtained in real municipal wastewater coagulated with the minimal doses of PIX (13.55 mg
Fe/L) and PAX (2.45 mg Al/L),PIX respectively [40]. PAX
It can be observed thatPIX these doses were PAXmuch higher
Properties
than the doses used in this sww/SD
study (3.02 mg Fe/L and
sww/SD 1.61 mg Al/L, respectively).
nww/SD nww/SD
At the lower
2 − 1 coagulant dose, larger size particles > 100 μm were prevalent as well. On the other
Sps (m ·g ) 355/13.9 283/10.0 283/10.0 350/12.2
hand, distributions
Dv10 (µm) of sludge particle classes 9.4/1.2
8.4/1.7 similar to the ones identified in sww7.4/1.2
9.4/1.2 have also been
recorded previously
Dv50 (µm) [39] in sludge
48.9/3.3 obtained from
99.9/4.5synthetic wastewater
99.9/4.5 coagulated with minimal
77.3/4.8
Dv90 (µm) 702/17.5 537/15.8 537/15.8 548/6.9
doses of PIX (6.5% of particles about 4 μm in size) and PAX (10% of particles about 10 μm in size).
Certain similarities and differences in particle size classes between sww and nww sludge discussed
hereTable
are most probably
1 presents a consequence
Sps and Dv (includingof very low doses
SD values) of coagulant;
for particles in this
of sludge study,ina the
obtained coagulant dose
conditions
was high in
illustrated enough
Figuresto2remove
and 5 andonlyspecified
50% of P.inDevelopment
Figures 6 andof 7. this issue will be completed by using the
values Sps and Dv, which are collected in Table 1. 2
The values of Sps (ranged between 283 and 355 m /g), were similar to the Sps values of
260–360 m2 /g, achieved for sludge obtained at coagulation of real municipal wastewater [40]. It
Table 1. The specific surface area (Sps) and volumetric dimension (Dv) values, for sludge particles
is note worthy that such Sps values were obtained for all doses (i.e., minimal, optimal and maximal) of
illustrated in Figure 8.
each coagulant PIX and PAX, respectively. Dv90 sww (Table 1), was slightly higher than Dv90 nww,
which means that the applied doses of both PIX coagulants
PAX formed slightly PAX
PIX more uniform/homogenous
Properties sww/SD sww/SD nww/SD nww/SD
particles in nww sludge than in sww sludge. It may also be suggested that the overall diversity of
structures of the particles Spsin sww
(m2∙gsludge
−1) is higher283/10.0
355/13.9 than the diversity particles in nww sludge.
283/10.0 of 350/12.2
Figure 9 contains a schematic
Dv10 (μm)presentation
8.4/1.7 of9.4/1.2
the development flocs in sww and nww sludge
9.4/1.2 of 7.4/1.2
obtained with PIX. The low Dv90 value for nww indicated a generally higher uniformity of flocs in
48.9/3.3 99.9/4.5 99.9/4.5
sludge built on sorg bridgesDv50 than(μm)
in sludge built on D bridges. The Dv90-based 77.3/4.8 floc uniformity concept
generally indicates the upper Dv90limit 702/17.5 537/15.8 537/15.8
(μm)for the diameter and thus, within the specified range, limits the
548/6.9
degree of raggedness or branching of these objects.
Table
Since 1 not
it is presents Spstask
an easy andto Dv (including
illustrate SD values)
this problem, for particles
the objects of sludge
presented obtained
in Figure in the
9 to a large
conditions
extent illustrated
are schematic in Figures 2 and
representations. 5 andlikely
It seems specified in Figures
that slightly 6 and
larger, 7.
negatively charged sorg bridges,
Thedirectly
anchored values of onSps
the(ranged
surface ofbetween
positive 283 and
prtc 355 effectively
more m2/g), wereclose
similar
the to the Sps
sludge values of
structures 260–360
than the
m2/g, achieved for sludge obtained at coagulation of real municipal wastewater [40]. It is note worthy
that such Sps values were obtained for all doses (i.e., minimal, optimal and maximal) of each
coagulant PIX and PAX, respectively. Dv90 sww (Table 1), was slightly higher than Dv90 nww, which
means that the applied doses of both coagulants formed slightly more uniform/homogenous particles
in nww sludge than in sww sludge. It may also be suggested that the overall diversity of structures of
the particles in sww sludge is higher than the diversity of particles in nww sludge.
sludge built on sorg bridges than in sludge built on D bridges. The Dv90-based floc uniformity
concept generally indicates the upper limit for the diameter and thus, within the specified range,
limits the degree of raggedness or branching of these objects.
Water Since it 101
2019, 11, is not an easy task to illustrate this problem, the objects presented in Figure 9 to a11large
of 15
extent are schematic representations. It seems likely that slightly larger, negatively charged sorg
bridges, anchored directly on the surface of positive prtc more−effectively close the sludge structures
smaller L bridges anchored on the prtc surface by the HPO , because L is rather unattractive for
than the smaller L bridges anchored on the prtc surface by the4HPO4−, because L is rather unattractive
positive prtc.
for positive prtc.
Thus, the left part of Figure 9 shows the closed, and the right part shows the partly open,
Thus, the left part of Figure 9 shows the closed, and the right part shows the partly open,
structure of two aggregates of nww sludge, similar in size, and an analogous, closed and completely
structure of two aggregates of nww sludge, similar in size, and an analogous, closed and completely
open structure of two aggregates of sww sludge. For such structures, Dv90 of nww would be lower
open structure of two aggregates of sww sludge. For such structures, Dv90 of nww would be lower
than Dv90 of sww, represented by the “open aggregate” in the lower, right-hand corner of Figure 9.
than Dv90 of sww, represented by the “open aggregate” in the lower, right-hand corner of Figure 9.
At the same time, the higher Dv90 value of sww sludge may generally indicate greater structural
At the same time, the higher Dv90 value of sww sludge may generally indicate greater structural
differentiation in sww sludge than in nww sludge.
differentiation in sww sludge than in nww sludge.

sorg
sorg
{Fe(OH)3}
sorg

{Fe(OH)3}

{Fe(OH)3}

{Fe(OH)3}
{Fe(OH) 3}

sorg
sorg

sorg
{Fe(OH)3} {Fe(OH)3}
sorg

L
{Fe(OH)3} {Fe(OH)3}

L L L
{Fe(OH)3}

L
{Fe(OH)3}

{Fe(OH)3}
L

L
L

L L
{Fe(OH)3}
{Fe(OH ) 3}
L
{Fe(OH)3}
L

Figure
Figure 9.
9. Schemas
Schemas ofof more
more homogenous
homogenous structures
structures of
of nww
nww sludge
sludge (Dv90
(Dv90 == 537
537 μm)
µm) and
and less
less
homogenous sww sludge (Dv90 = 702 μm) obtained after coagulation with PIX.
homogenous sww sludge (Dv90 = 702 µm) obtained after coagulation with PIX.

Unquestionably, the progress of aaf processes processes is is a direct consequence

consequence of the primary process
forming
forming the theunits
unitsillustrated
illustratedin in Figures
Figures 3 and 4. Due
3 and 4. to Duea large
to a deficit of the coagulant
large deficit (an amount
of the coagulant (an
needed
amountto remove
needed to just
remove50% just
of P),50%theofconditions applied in
P), the conditions this study
applied canstudy
in this be referred
can be to as “sub-
referred to
stoichiometric”. In line with
as “sub-stoichiometric”. thewith
In line Langmuir’s theory oftheory
the Langmuir’s adsorption, under such
of adsorption, underconditions we should
such conditions we
expect
shouldcomplete saturation/use
expect complete of the surface
saturation/use of prtc by
of the surface of adsorbed substances;
prtc by adsorbed i.e., L and
substances; i.e.,P-PO
L and4 in sww4
P-PO
flocs,
in swwand SS,and
flocs, sorgSS,
andsorgP and
in nwwP inflocs. In thisIncase,
nww flocs. individual
this case, unitsunits
individual of flocs being
of flocs formed
being should
formed be
should
characterized
be characterized by by
higher
higher porosity
porosity than
thanunits
unitsformed
formedunderunder“stoichiometric”
“stoichiometric”conditions,
conditions,especially
especially in
in
excessive
excessive amounts
amounts of of aa coagulant
coagulant [37,39].
[37,39]. Porosity
Porosity ofof floc-cluster
floc-cluster units
units may
may influence
influence thethe growth
growth of
Sps. At
At the
the same
same time,
time, thethe larger
larger filling
filling of
of the
the centers
centers onon the
the adsorbent
adsorbent surface
surface (prtc)
(prtc) should
should favor the
aaf processes,
processes, as
asthe
theforces
forcesof ofmutual
mutualrepulsion
repulsiondecrease
decreasebetween
betweenthethe“saturated”
“saturated”surfaces
surfacesof ofthe
theunits.
units.
As a result, the final porosity of these units only exerts a slight effect on the size, and on the final Sps
value, for closed floc structures formed under sub-stoichiometric conditions.
It has been explained [43] that flocs of the analyzed sludge are formed from the units presented in
Figure 9. Filling the space with the mass of a sludge floc can be defined with the fractal dimension,
Df , previously defined in Equation (2). There are more data available [19,29,46–48] on the fractal
Water 2019, 11, 101 12 of 15

dimension values of various types of flocs, including the values of Df in sludge of municipal wastewater.
Depending on the measurement method applied, the type of a coagulant and other parameters of the
process, the values of Df can be equal with the following values: 1.72 [19]; 1.68–1.74 [3]; 1.50–1.87 [4];
1.69–1.96 [20]; 1.7–1.8 [46], 1.67–1.90 [47] and 1.8 [48]. Both fundamentally and practically, it may be
interesting to compare the number of units, schematically illustrated in Figures 3 and 4, in a single
floc of sww and nww sludge. Based on the aforementioned Df values, the following assumptions were
made for simple calculations of the number of units in such “statistical” floc: Df = 1.75 and R = Dv50.
For comparative purposes, it seems sufficient to choose R = Dv50, as it is implied by the definition of
Dv, according to which this value can be the closest to the R diameter of a “statistical” floc. The values
of the number of n units in a statistical municipal wastewater sludge floc, as comprised in Table 2,
were calculated from the following equation:

n = [ R/r ] D f (8)

where: R = Dv50 in µm, while r = 0.17 µm and is an approximate sum of diameters of individual
components in a sww unit:

r(H2 PO4 − ) + r{Al(OH)3 } + r(H2 PO4 − ) + r(D)

(9)
= 0.0012 + 0.165 + 0.0012 + 0.010 = 0.1685 (µm)

Earlier, an assumption was made that r(sorg) is slightly larger than r(D). Simultaneously,
considering the fact that the r of a unit of nww sludge:

r{Al(OH)3 } + r(sorg) (10)

does not contain r(H2 PO4 − ) and in view of the lack of other data for calculations, the value of r for
nww was also adopted as being equal to 0.17 µm, because the dominant component of r of a nww floc
always has to be r{Al(OH)3 }.

Table 2. Number of units in a statistical floc of the analyzed sludge.

Coagulant/Wastewater Type Dv50 (µm) Number of Units in a Floc

PIX/sww 48.9 20,091
PAX/sww 32.4 9776
PIX/nww 99.9 70,138
PAX/nww 77.3 44,774

The data contained in Table 2 are not absolute values and can be used only for comparative
purposes. They indicate a much denser filling of the space by units-clusters in a bigger floc (Dv50) of
nnw sludge than that of a smaller floc (Dv50) in sww sludge. Within a certain range, these data confirm
the general hypothesis of higher homogeneity of nww sludge, than of sww sludge (Figure 9).
The data allowed a comparison of the structures of two types of municipal wastewater sludge.
In all presented case studies, the obtained sludge after chemical treatment of wastewater contained
3–5% dry solids, of which an average of 45% of the dry solids content was organic.
The results of such studies, when advanced and developed, can aid the improvement of practical
wastewater treatment technologies. Simple procedures of simultaneous online measurements of Dv10,
Dv50, Dv90 and Sps could be easily implemented at wastewater treatment plants. If a wastewater
treatment plant is equipped with such an online measuring system, it will allow:

(1) additional monitoring of the coagulation-flocculation process,

(2) broader assessment of the “health”/quality of activated sludge,
(3) better control and regulation of the process of municipal wastewater sludge dewatering.
Water 2019, 11, 101 13 of 15

4. Conclusions
Molecules of dye are adsorbed to prtc by bridging, with the help of H2 PO4 − ions previously
adsorbed to these surfaces. Unlike in sww, negatively charged sorg in nww are directly adsorbed on the
surface of prtc.
As the dose of a coagulant increased, the CSTsww increased, while the CSTnww decreased. The sww
sludge with PAX binds water more effectively at higher Al doses. Aggregates composed of spherical
{Al(OH)3 } units leave more internal space for water than aggregates built from rod-shaped {Fe(OH)3 }
units and, in consequence, a higher dose of PAX (Al) means that it is more difficult to remove water
from Al-sww sludge (longer CST) than from Fe-sww (shorter CST).
Particles that are smaller in size dominate in sww sludge, while larger particles prevail in nww
sludge. Except for PIX in nww, the analyzed particles of the types of sludge tested were characterized
by similar Sps, irrespective of the type of wastewater or the applied coagulant. Values of Dv90 for
sww are slightly higher than Dv90 nww, which means that the differentiation of structures of sww
sludge particles is greater than in nww sludge. Slightly larger, negatively charged sorg bridges, directly
anchored on the surface of positive prtc, more effectively close structures of nww sludge than smaller
L bridges, anchored on the surface of prtc via HPO4 − ions. The porosity of units has only a slight
influence on the size and final value of Sps of closed structures of flocs formed under sub-stoichiometric
conditions. The larger statistical floc of nww sludge is more densely packed with units-clusters than
the space of a smaller sww floc.

Author Contributions: This article was written by L.S. and H.R. based on the investigations carried out in the
frame of Polish-Norwegian Research Program operated by the National Centre for Research and Development
under the Norwegian Financial Mechanism 2009–2014. These authors designed the research project and conducted
the laboratory tests and analyses developed by Ph.D students M.K. and M.T. S.K. designed the equipment and
provided technical knowledge to support the experimental work. M.S. carried out optical investigations and
visualization and I.C. participated in reviewing and final writing of this manuscript.
Funding: This research was mainly funded by Polish-Norwegian Research Program, grant number,
POL-NOR/196364/7/2013.; 561755-1-2015-1-NO-EPPKA2-CBHE-JP, (2015-3386/001-001); and statutory
grant 20610.001-300.
Acknowledgments: This work was supported by a Polish-Norwegian Research Program operated by the National
Centre for Research and Development under the Norwegian Financial Mechanism 2009–2014 in the framework of
Project Contract No. POL-NOR/196364/7/2013; “Harmonizing water related graduate education” under grant
561755-1-2015-1-NO-EPPKA2-CBHE-JP (2015-3386/001-001); and under statutory grant 20610.001-300.
Conflicts of Interest: The authors declare no conflicts of interest.

Nomenclature
sww synthetic wastewater
nww real municipal wastewater (natural origin)
CST capillary suction time (s)
aaf aggregation-agglomeration-flocculation
SD standard deviation of the sample (% or mg/L)
fes number of mg Fe3+ (PIX) which remove 50% P from 1 L of sww
als umber of mg Al3+ (PAX) which remove 50% P from 1 L of sww
fen number of mg Fe3+ (PIX) which remove 50% P from 1 L of nww
aln number of mg Fe3+ (PIX) which remove 50% P from 1 L of nww
Dv volumetric dimension (µm)
Sps specific surface area (m2 /g)
L disperse dye (Synthene Scarlet P3GL)
SS Suspended Solids (mg/L)
prtc colloidal particle {Fe(OH)3 } or {Al(OH)3 }
sorg organic substances (responsible for COD in nww)
Water 2019, 11, 101 14 of 15

References
1. Duan, J.; Gregory, J. Coagulation by hydrolyzing metal salts. Adv Colloid Interface Sci. 2003, 100–102, 475–502.
[CrossRef]
2. The, C.Y.; Budiman, P.M.; Shak, K.P.; Wu, T.Y. Recent advancement of coagulation-flocculation and its
application in wastewater treatment. Ind. Eng. Chem. Res. 2016, 55, 4363–4389.
3. Zheng, H.; Shu, G.; Jiang, S.; Tshukudu, T.; Xiang, X.; Zhang, P.; He, Q. Investigations of
coagulation-flocculation process by performance optimization, model prediction and fractal structure
of flocs. Desalination 2011, 269, 148–156. [CrossRef]
4. Chu, C.P.; Lee, D.J. Effect of pre-hydrolysis on floc structure. J. Environ. Manag. 2004, 71, 285–292. [CrossRef]
5. Zhao, Y.X.; Shon, H.K.; Wang, Y.; Kim, J.H.; Yue, Q.Y. The effect of second coagulant dose on the regrowth of
flocs formed by charge neutralization and sweep coagulation using titanium tetrachloride (TiCl4 ). J. Hazard
Mater. 2011, 198, 70–77. [CrossRef]
6. Fytili, D.; Zabaniotou, A. Utilization of sewage sludge in EU application of old and new methods—A review.
Renew. Sustain. Energy Rev. 2008, 12, 116–140. [CrossRef]
7. Verna, S.; Prased, B.; Mishra, I.M. Pretreatment of petrochemical wastewater by coagulation and the sludge
characteristics. J. Hazard Mater. 2010, 178, 1055–1064. [CrossRef]
8. Rohrsetzer, S.; Paszli, I.; Csempesz, F.; Ban, S. Colloidal stability of electrostatically stabilized sol particles.
Part I: The role of hydration in coagulation and repeptization of ferric hydroxide sol. Colloid Polym. Sci. 1992,
270, 1243–1251. [CrossRef]
9. Smoczyński, L.; Ratnaweera, H.; Kosobucka, M.; Kvaal, K.; Smoczyński, M. Image Analysis of Sludge
Aggregates Obtained at Preliminary Treatment of Sewage. Water Sci. Technol. 2014, 70, 1048–1055. [CrossRef]
10. Amirtharajah, A.; Mills, M.K. Rapid-mix design for mechanism of alum coagulation. JAWWA 1982, 74,
210–216. [CrossRef]
11. Butler, E.; Hung, Y.T.; Yu, L.R.; Al Ahmad, M. Electrocoagulation in wastewater treatment. Water 2011, 3,
395–525. [CrossRef]
12. Zaleschi, L.; Teodosiu, C.; Cretescu, I.; Rodrigo, M.A. A Comparative Study of Electrocoagulation and
Chemical Coagulation Processes Applied for Wastewater Treatment. Environ. Eng. Manag. J. 2012, 11,
1517–1525.
13. Smoczynski, L.; Munska, K.; Pierozynski, B. Electrocoagulation of synthetic dairy wastewater. Water Sci.
Technol. 2013, 67, 404–409. [CrossRef]
14. Groterud, O.; Smoczyński, L. Removal of phosphorus and residual aluminium by a recirculating electrolysis
of wastewater. Vatten 1986, 42, 293–296.
15. Mikkelsen, L.H.; Keiding, K. Physico-chemical characteristics of full scale sewage sludge with implications
to dewatering. Water Res. 2002, 36, 2451–2462. [CrossRef]
16. Jin, B.; Wilen, B.M.; Lant, P. Impact of morphological, physical and chemical properties of sludge flocs on
dewaterability of activated sludge. Chem. Eng. J. 2004, 98, 115–126. [CrossRef]
17. Turchiuli, C.; Fargues, C. Influence of structural properties of alum and ferric flocs on sludge dewaterability.
Chem. Eng. J. 2004, 103, 123–131. [CrossRef]
18. Niu, M.; Zhang, W.; Wang, D.; Chen, Y.; Chen, R. Correlation of physiochemical properties and sludge
dewaterability under chemical conditioning using inorganic coagulants. Bioresour. Technol. 2013, 144, 337–343.
[CrossRef]
19. Zhao, Y.Q. Correlations between floc physical properties and optimum polymer dosage in alum sludge
conditioning and dewatering. Chem. Eng. J. 2003, 92, 227–235. [CrossRef]
20. Zhao, P.; Ge, S.; Chen, Z.; Li, X. Study on pore characteristics of flocs and sludge dewaterability based on
fractal methods (pore characteristics of flocs and sludge dewatering). Appl. Therm. Eng. 2013, 58, 217–223.
[CrossRef]
21. Sawalha, O.; Scholz, M. Assessment of capillary suction time (CST) test methodology. Environ. Technol. 2007,
28, 1377–1386. [CrossRef]
22. Sawalha, O.; Scholz, M. Innovative enhancement of the design and precision of the capillary suction time
testing device. Water Environ. Res. 2009, 81, 2344–2352. [CrossRef]
23. Scholtz, M.; Tapp, J. Development of the revised Capillary Suction Time (CST) test. Water Cond. Purif. 2006,
48, 46–52.
Water 2019, 11, 101 15 of 15

24. Tuan, P.-A.; Sillanpaa, M. Migration of ions and organic matter during electro-dewatering of anaerobic
sludge. J. Hazard Mater. 2010, 173, 54–61. [CrossRef]
25. Lee, C.H.; Liu, J.C. Sludge dewaterability and floc structure in dual polymer conditioning. Adv. Environ. Res.
2001, 5, 129–136. [CrossRef]
26. Meakin, P. The effects of reorganization processes on two dimensional cluster—Cluster aggregation. J. Colloid
Interface Sci. 1986, 112, 187–194. [CrossRef]
27. Yu, W.; Liu, T.; Gregory, J.; Li, G.; Liu, H.; Qua, J. Aggregation of nano-sized alum–humic primary particles.
Sep. Purif. Technol. 2012, 99, 44–49. [CrossRef]
28. Pastor-Satorras, R.; Rubí, J.M. Particle-cluster aggregation with dipolar interactions. Phys. Rev. E 1995, 51,
5994–6003. [CrossRef]
29. Smoczyński, L.; Bukowski, Z.; Wardzyńska, R.; Zał˛eska-Chróst, K. Dłużyńska, B. Simulation of coagulation,
flocculation and sedimentation. Water Environ. Res. 2009, 81, 348–356. [CrossRef]
30. Jullien, R.; Meakin, P. Simple models for the restructuring of 3-dimensional ballistic aggregates. J. Colloid
Interface Sci. 1989, 127, 265–272. [CrossRef]
31. Jarvis, P.; Jefferson, B.; Parsons, S.A. Measuring floc structural characteristics. Rev. Environ. Sci. Bio-Technol.
2005, 4, 1–18. [CrossRef]
32. Chu, C.P.; Lee, D.J. Structural analysis of sludge flocs. Adv. Powder Technol. 2004, 15, 515–532. [CrossRef]
33. Zhao, Y.Q. Settling behaviour of polymer flocculated water treatment sludge II: Effects of floc structure and
floc packing. Sep. Purif. Technol. 2004, 35, 175–183. [CrossRef]
34. Mandelbrot, B.B. The Fractal Geometry of Nature; Freeman: San Francisco, CA, USA, 1982.
35. Leman, J.; Smoczynski, M.; Dolgan, T.; Dziuba, Z. Fractal analysis of structure of cow and goat β-lactoglobulin
preparation. J. Food Technol. 2005, 42, 428–430.
36. Yang, Z.; Yang, H.; Jiang, Z.; Huang, X.; Li, H.; Li, A.; Cheng, R. A new method for calculation of flocculation
kinetics combining Smoluchowski model with fractal theory. Colloids Surf. A 2013, 423, 11–19. [CrossRef]
37. Meng, F.G.; Zhang, H.M.; Li, Y.S.; Zhang, X.W.; Yang, F.L.; Xiao, J.N. Cake layer morphology in microfiltration
of activated sludge wastewater based on fractal analysis. Sep. Purif. Technol. 2005, 44, 250–257. [CrossRef]
38. Smoczynski, L.; Ratnaweera, H.; Kosobucka, M.; Smoczynski, M.; Pieczulis-Smoczynska, K.; Cretescu, I. The
size of aggregates formed during coagulation and electrocoagulation of synthetic wastewater. J. Environ.
Prot. Ecol. 2016, 17, 1160–1170.
39. Wang, Y.; Gao, B.Y.; Xu, X.M.; Xu, G.Y. Characterization of floc size, strength and structure in various
aluminum coagulants treatment. J. Colloid Interface Sci. 2009, 332, 354–359. [CrossRef]
40. Smoczynski, L.; Kosobucka, M.; Smoczynski, M.; Ratnaweera, H.; Pieczulis-Smoczynska, K. Sizes of particles
formed during municipal wastewater treatment. Water Sci. Technol. 2017, 75, 971–977.
41. Smoczyński, L.; Ratnaweera, H.; Kosobucka, M.; Smoczyński, M.; Kalinowski, S.; Kvaal, K. Modelling the
structure of sludge aggregates. Environ. Technol. 2016, 37, 1122–1132. [CrossRef]
42. Smoczyński, L.; Ratnaweera, H.; Kosobucka, M.; Smoczyński, M. Image analysis of sludge aggregates. Sep.
Purif. Technol. 2014, 122, 412–420. [CrossRef]
43. Malvern Instruments Limited. A Basic Guide to Particle Characterization; Malvern Instruments Limited:
Worcestershire, UK, 2012; pp. 1–26.
44. Macedo, M.L.F.; Osawa, C.C.; Bertran, C.A. Sol-gel synthesis of transparent alumina gel and pure gamma
alumina by urea hydrolysis of aluminum nitrate. J. Sol.-Gel. Sci. Technol. 2004, 30, 135–140. [CrossRef]
45. Haas, W.; Zrinyi, M.; Kilian, H.G.; Heise, B. Structural analysis of anisometric colloidal iron(III)-hydroxide
particles and particle-aggregates incorporated in poly(vinyl-acetate) networks. Colloid Polym. Sci. 1993, 271,
1024–1034. [CrossRef]
46. Waite, T.D. Measurement and implications of flock structure in water and wastewater treatment. Colloids
Surf. A 1999, 151, 27–41. [CrossRef]
47. Smoczynski, L.; Wardzynska, R. Study on macroscopic aggregation of silica suspensions and sewage.
J. Colloid Interface Sci. 1996, 183, 309–314. [CrossRef]
48. Gregory, J. The role of floc density in solid-liquid separation. Filtr. Sep. 1998, 35, 367–371. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).